CN104102816A - Symptom match and machine learning-based automatic diagnosis system and method - Google Patents

Abstract

The invention discloses a symptom match and machine learning-based automatic diagnosis system and method. The system comprises a disease/symptom database for storing each known disease/symptom and the corresponding symptoms thereof, a user interaction module for receiving a symptom keyword set input by the user, a symptom match module for matching the symptoms in the disease/symptom database according to the symptom keyword set input by the user and calculating the match degree between the symptom keyword set and each disease/symptom, and a diagnosis module for determining the corresponding disease/symptom according to the match degree between the symptom keyword set and each disease/symptom.

Description

Based on auto-check system and the method for symptom coupling and machine learning

Technical field

The present invention relates to medical information field, in particular to a kind of auto-check system and method based on symptom coupling and machine learning.

Background technology

The medical terminology of using in following paper the present invention:

Disease: be that the perverse trend of causing a disease acts on human body, human righteousness resists with it and the body imbalance of yin and yang that causes, internal organs damage, physiological function are not normal or a complete life process of psychological activity obstacle.

Disease: the pathology that is certain one-phase or a certain type in lysis is summarized, generally have one group relatively-stationary, have sings and symptoms inner link, that can disclose disease one-phase or a certain type pathology essence to form.

Symptom: being indivedual, the isolated phenomenon showing in lysis, can be subjective sensation that patient is abnormal or behavior performance can be also the abnormal sign that doctor finds while checking patient.

Along with the raising day by day of the level of informatization, people can obtain medical information by various information terminals, but how according to known symptom offer user accurately disease/disease diagnostic result be still a problem of needing solution badly.

Summary of the invention

The invention provides a kind of auto-check system and method based on symptom coupling and machine learning, in order to offer user's disease/disease diagnostic result accurately according to known symptom.

For achieving the above object, the invention provides a kind of auto-check system based on symptom coupling and machine learning, comprising:

Disease/disease database, for preserving known every kind of disease/disease and corresponding symptom thereof;

User interactive module, for receiving the symptom keyword set of user's input;

Symptom matching module, for mating with the symptom of described disease/disease database according to the described symptom keyword set of user's input, calculates the matching degree of described symptom keyword set and every kind of disease/disease;

Diagnostic module, for determining corresponding disease/disease according to described symptom keyword set with the matching degree of every kind of disease/disease.

Further, said system also comprises:

Vocabulary builds module, for building symptom degree of correlation vocabulary, is specially:

Obtain symptom data, wherein said symptom data comprise the symptom obtained from textbook, dictionary with the symptom set of, near synonym table, every disease/disease obtaining from described disease/disease database and ask every effectively symptom set of request of obtaining record from user;

For obtained symptom data, suppose to have two symptom x and y, the degree of association μ (x, y) of these two symptom x and y is

$\mathrm{\μ}(x,y)={\mathrm{\Σ}}_P[\mathrm{\ρ}\left(P\right)\·\frac{{\mathrm{\Σ}}_{r\∈P}g(r,x,y)}{{\mathrm{\Σ}}_{p\∈P}f(p,x)\·{\mathrm{\Σ}}_{q\∈P}f(q,y)}]$

Wherein ρ (P) represents the judgement weight of data source P, artificially sets ρ (with near synonym table) > ρ (disease/syndrome storehouse) >=ρ (user asks record) according to expertise; R, p, q represents the each syndrome set in data source P;

wherein | p| represents the number of the symptom containing in symptom set p,

Two symptoms that the degree of association is greater than to degree of association threshold value are saved in the symptom degree of correlation vocabulary of establishment.

Further, described symptom matching module comprises:

Weight calculation unit, for calculate the weights W (d, x) of symptom x at disease/disease d according to following formula:

Wherein, ρ (S) represents the weight of data source S, and e represents each description unit information of diseases related/disease in data source S;

Matching degree computing unit, for calculating every disease/disease of described disease/disease database with respect to the matching degree of described symptom keyword set, is specially:

Suppose that the described symptom keyword set that user provides is combined into A, travel through each disease/disease d and corresponding symptom set σ (d) thereof in described disease/disease database;

Matching degree M (A, d) with following formula calculating disease/disease d with respect to described symptom keyword set A:

$M(A,d)=\frac{{\mathrm{\Σ}}_{x\∈A,y\∈\mathrm{\σ}\left(d\right)}\mathrm{\μ}(x,y)\·W(d,y)}{\left|A\right|\·|\mathrm{\σ}\left(d\right)|};$Wherein, | A| and | σ (d) | represent respectively the element number in set A and set σ (d);

Press M (A, d) descending order sorts to corresponding disease/disease, the result that sequence is obtained represents with R and presents to user, wherein R={d|M (A, d) > 0 and r (d) < N}, r (d) represents press the sequence number of disease/disease corresponding after the descending sequence of M (A, d), and N is the constant of artificial setting.

Further, said system also comprises:

Update module, for supplementing and upgrading described disease/disease database.

Further, the symptom set of described effective request refers in the matching result of this request and contains disease/disease that matching degree is greater than the constant C of setting.

For achieving the above object, the present invention also provides a kind of automatic diagnosis method based on symptom coupling and machine learning, comprises the following steps:

Receive the symptom keyword set of user's input;

Mate with the symptom in disease/disease database according to the symptom keyword set of user's input, calculate the matching degree of described symptom keyword set and every kind of disease/disease, wherein said disease/disease database is preserved known every kind of disease/disease and corresponding symptom thereof;

Determine corresponding disease/disease according to described symptom keyword set with the matching degree of every kind of disease/disease.

Further, further comprising the steps of before mating step according to the symptom keyword set of user's input with the symptom in disease/disease database:

Build symptom degree of correlation vocabulary, specifically comprise:

Obtain symptom data, wherein said symptom data comprise the symptom obtained from textbook, dictionary with the symptom set of, near synonym table, every disease/disease obtaining from described disease/disease database and ask every effectively symptom set of request of obtaining record from user;

For obtained symptom data, suppose to have two symptom x and y, the degree of association μ (x, y) of these two symptom x and y is

$\mathrm{\&mu;}(x,y)={\mathrm{\&Sigma;}}_P[\mathrm{\&rho;}\left(P\right)\&CenterDot;\frac{{\mathrm{\&Sigma;}}_{r\&Element;P}g(r,x,y)}{{\mathrm{\&Sigma;}}_{p\&Element;P}f(p,x)\&CenterDot;{\mathrm{\&Sigma;}}_{q\&Element;P}f(q,y)}]$

ρ (P) represents the judgement weight of data source P, artificially sets ρ (with near synonym table) > ρ (disease/syndrome storehouse) >=ρ (user asks record) according to expertise; R, p, q represents the each syndrome set in data source P;

wherein | p| represents the number of the symptom containing in symptom set p,

Two symptoms that the degree of association is greater than to degree of association threshold value are saved in the symptom degree of correlation vocabulary of establishment.

Further, the described symptom keyword set according to user's input is mated with the symptom in disease/disease database, and the matching degree step of calculating described symptom keyword set and every kind of disease/disease comprises:

Calculating the weights W (d, x) of symptom x in disease/disease d is

Wherein, ρ (S) represents the weight of data source S,

e represents each description unit information of diseases related/disease in data source S;

Suppose that the described symptom keyword set that user provides is combined into A, travel through each disease/disease d and corresponding symptom set σ (d) thereof in described disease/disease database;

Matching degree M (A, d) with following formula calculating disease/disease d with respect to described symptom keyword set A

$M(A,d)=\frac{{\mathrm{\&Sigma;}}_{x\&Element;A,y\&Element;\mathrm{\&sigma;}\left(d\right)}\mathrm{\&mu;}(x,y)\&CenterDot;W(d,y)}{\left|A\right|\&CenterDot;|\mathrm{\&sigma;}\left(d\right)|};$

Wherein, | A| and | σ (d) | represent respectively the element number in set A and set σ (d);

Press M (A, d) descending order sorts to corresponding disease/disease, the result that sequence is obtained represents with R and presents to user, wherein R={d|M (A, d) > 0 and r (d) < N}, r (d) represents press the sequence number of disease/disease corresponding after the descending sequence of M (A, d), and N is the constant of artificial setting.

Further, said method is further comprising the steps of:

Described disease/disease database is supplemented and upgraded.

Further, the symptom set of described effective request refers in the matching result of this request and contains disease/disease that matching degree is greater than the constant C of setting.

One group of symptom symptom corresponding with the disease of having included in system and syndrome that the present invention provides user is mated, automatically infer and cause possible disease and the syndrome of this group symptom by calculating matching degree, thereby provide relatively accurate diagnostic result to user.

Brief description of the drawings

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

Fig. 1 is the auto-check system module map based on symptom coupling and machine learning of one embodiment of the invention;

Fig. 2 is the auto-check system fundamental diagram based on symptom coupling and machine learning of a preferred embodiment of the invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiment.Based on the embodiment in the present invention, those of ordinary skill in the art, not paying the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.

Fig. 1 is the auto-check system module map based on symptom coupling and machine learning of one embodiment of the invention; Fig. 2 is the auto-check system fundamental diagram based on symptom coupling and machine learning of a preferred embodiment of the invention.As shown in the figure, this auto-check system comprises:

Disease/disease database, for preserving known every kind of disease/disease and corresponding symptom thereof;

Wherein, in the time building disease/disease database, selected data source can be national standard (as " tcm clinical practice diagnosis and treatment term disease part "), modern Chinese medicine textbook, modern Chinese medicine dictionary etc., illness, case discussion etc. in traditional Chinese medical science ancient books, and Modern Medical Record data.

User interactive module, for receiving the symptom keyword set of user's input;

Symptom matching module, for mating with the symptom of described disease/disease database according to the described symptom keyword set of user's input, calculates the matching degree of described symptom keyword set and every kind of disease/disease;

Diagnostic module, for determining corresponding disease/disease according to described symptom keyword set with the matching degree of every kind of disease/disease.

Further, said system also comprises:

Vocabulary builds module, for building symptom degree of correlation vocabulary, is specially:

Obtain symptom data, wherein said symptom data comprise the symptom obtained from textbook, dictionary with the symptom set of, near synonym table, every disease/disease obtaining from described disease/disease database and ask every effectively symptom set of request of obtaining record from user; The textbook here, dictionary can be for example traditional Chinese medical science voluminous dictionary, tcm symptom antidiastole, Chinese near synonym dictionary etc., and the symptom set of effective request here refers in the matching result of this request and contains disease/disease that matching degree is greater than the constant C of setting;

For obtained symptom data, suppose to have two symptom x and y, the degree of association μ (x, y) of these two symptom x and y is

$\mathrm{\&mu;}(x,y)={\mathrm{\&Sigma;}}_P[\mathrm{\&rho;}\left(P\right)\&CenterDot;\frac{{\mathrm{\&Sigma;}}_{r\&Element;P}g(r,x,y)}{{\mathrm{\&Sigma;}}_{p\&Element;P}f(p,x)\&CenterDot;{\mathrm{\&Sigma;}}_{q\&Element;P}f(q,y)}]$

Wherein ρ (P) represents the judgement weight of data source P, artificially sets ρ (with near synonym table) > ρ (disease/syndrome storehouse) >=ρ (user asks record) according to expertise; R, p, q represents the each syndrome set in data source P;

wherein | p| represents the number of the symptom containing in symptom set p,

Two symptoms that the degree of association is greater than to degree of association threshold value are saved in the symptom degree of correlation vocabulary of establishment.

Further, described symptom matching module comprises:

Weight calculation unit, for calculate the weights W (d, x) of symptom x at disease/disease d according to following formula:

Wherein, ρ (S) represents the weight of data source S, artificially set according to expertise, meet ρ (national standard) > ρ (textbook, dictionary) >=ρ (traditional Chinese medical science ancient books) > ρ (modern case, e represents each description unit information of diseases related/disease in data source S;

Matching degree computing unit, for calculating every disease/disease of described disease/disease database with respect to the matching degree of described symptom keyword set, is specially:

Suppose that the described symptom keyword set that user provides is combined into A, travel through each disease/disease d and corresponding symptom set σ (d) thereof in described disease/disease database;

Matching degree M (A, d) with following formula calculating disease/disease d with respect to described symptom keyword set A:

$M(A,d)=\frac{{\mathrm{\&Sigma;}}_{x\&Element;A,y\&Element;\mathrm{\&sigma;}\left(d\right)}\mathrm{\&mu;}(x,y)\&CenterDot;W(d,y)}{\left|A\right|\&CenterDot;|\mathrm{\&sigma;}\left(d\right)|};$

Wherein, | A| and | σ (d) | represent respectively the element number in set A and set σ (d);

Press M (A, d) descending order sorts to corresponding disease/disease, the result that sequence is obtained represents with R and presents to user, wherein R={d|M (A, d) > 0 and r (d) < N}, r (d) represents press the sequence number of disease/disease corresponding after the descending sequence of M (A, d), and N is the constant of artificial setting.

Further, said system also comprises:

Update module, for supplementing and upgrading described disease/disease database.

Wherein, can be regularly the opportunity of renewal, as carried out once weekly; Also can be the immediate updating being triggered by accident, as new national standard is promulgated, World Health Organization announces new pandemic information etc.

Example:

User's input: sympotomatic set A: diarrhoea, stomach-ache, anus heat, slippery and rapid pulse

Disease/the disease of including

D₁: the damp and hot wound of having loose bowels

σ (d₁): the stomachache of having loose bowels, rush down that lower urgent, stool being yellowish-brown, burning sensation of the anus, dysphoria with smothery sensation are thirsty, oliguria with yellow urine, slippery and rapid pulse

D₂: cholera humidifier fever

σ (d₂): if the profit of telling, fever and chills bodily pain, depressed, the n and V of fever and chills abdominal pain, ambition, active borhorygmus are not had a pain, are gushed soft stool around navel and continue and rush down clear water or swill juice, peripheral coldness, lower leg contraction, slippery and rapid pulse

Suppose according to data with existing source, by μ (x, y) formula calculates μ (diarrhoea, the stomachache of having loose bowels)=0.14, μ (diarrhoea is rushed down lower urgent)=0.43, μ (stomach-ache, the stomachache of having loose bowels)=0.31, μ (anus heat, burning sensation of the anus)=0.61, μ (slippery and rapid pulse, slippery and rapid pulse)=1.00, μ (stomach-ache, active borhorygmus is had a pain around navel)=0.14, μ (diarrhoea, if gush soft stool continue rush down clear water or swill juice)=0.07, other μ (x, y) is 0;

And according to data with existing source, calculate W (d by the formula of W (d, y)₁, the stomachache of having loose bowels)=0.71, W (d₁, rush down lower urgent)=0.65, W (d₁, burning sensation of the anus)=0.57, W (d₁, slippery and rapid pulse)=0.41, W (d₂, active borhorygmus is had a pain around navel)=0.31, W (d₂if, gush soft stool continue rush down clear water or swill juice)=0.57,, other W (d, y) is 0;

Can calculate d by the formula of M (A, d)₁and d₂as follows with the matching degree of the symptom set of user input respectively:

M (A, d₁μ)=[(diarrhoea, the stomachache of having loose bowels) × W (d₁, the stomachache of having loose bowels) and+μ (diarrhoea is rushed down lower urgent) × W (d₁, rush down lower urgent) and+μ (stomach-ache, the stomachache of having loose bowels) × W (d₁, the stomachache of having loose bowels) and+μ (anus heat, burning sensation of the anus) × W (d₁, burning sensation of the anus) and+μ (slippery and rapid pulse, slippery and rapid pulse) × W (d₁, slippery and rapid pulse)]/(| A|| σ (d₁) |)=0.0485

M (A, d₂μ)=[(stomach-ache, active borhorygmus is had a pain around navel) × W (d₂, active borhorygmus is had a pain around navel)+μ (diarrhoea, if gush soft stool continue rush down clear water or swill juice) × W (d₂if, gush soft stool continue rush down clear water or swill juice)]/(| A|| σ (d₂) |)=0.0021

Due to M (A, d₁) > M (A, d₂), therefore in the diagnostic result R that returns to user, M (A, d₁) be placed on before.Embodiment adapts with said system, is below the automatic diagnosis method embodiment based on symptom coupling and machine learning of one embodiment of the invention, comprises the following steps:

Receive the symptom keyword set of user's input;

Mate with the symptom in disease/disease database according to the symptom keyword set of user's input, calculate the matching degree of described symptom keyword set and every kind of disease/disease, wherein said disease/disease database is preserved known every kind of disease/disease and corresponding symptom thereof;

Determine corresponding disease/disease according to described symptom keyword set with the matching degree of every kind of disease/disease.

Further, further comprising the steps of before mating step according to the symptom keyword set of user's input with the symptom in disease/disease database:

Build symptom degree of correlation vocabulary, specifically comprise:

Obtain symptom data, wherein said symptom data comprise the symptom obtained from textbook, dictionary with the symptom set of, near synonym table, every disease/disease obtaining from described disease/disease database and ask every effectively symptom set of request of obtaining record from user;

For obtained symptom data, suppose to have two symptom x and y, the degree of association μ (x, y) of these two symptom x and y is

$\mathrm{\&mu;}(x,y)={\mathrm{\&Sigma;}}_P[\mathrm{\&rho;}\left(P\right)\&CenterDot;\frac{{\mathrm{\&Sigma;}}_{r\&Element;P}g(r,x,y)}{{\mathrm{\&Sigma;}}_{p\&Element;P}f(p,x)\&CenterDot;{\mathrm{\&Sigma;}}_{q\&Element;P}f(q,y)}]$

ρ (P) represents the judgement weight of data source P, wherein

Two symptoms that the degree of association is greater than to degree of association threshold value are saved in the symptom degree of correlation vocabulary of establishment.

Further, the described symptom keyword set according to user's input is mated with the symptom in disease/disease database, and the matching degree step of calculating described symptom keyword set and every kind of disease/disease comprises:

Calculating the weights W (d, x) of symptom x in disease/disease d is

Wherein, ρ (S) represents the weight of data source S, and e represents each description unit information of diseases related/disease in data source S;

Suppose that the described symptom keyword set that user provides is combined into A, travel through each disease/disease d and corresponding symptom set σ (d) thereof in described disease/disease database;

Matching degree M (A, d) with following formula calculating disease/disease d with respect to described symptom keyword set A

$M(A,d)=\frac{{\mathrm{\&Sigma;}}_{x\&Element;A,y\&Element;\mathrm{\&sigma;}\left(d\right)}\mathrm{\&mu;}(x,y)\&CenterDot;W(d,y)}{\left|A\right|\&CenterDot;|\mathrm{\&sigma;}\left(d\right)|};$Press M (A, d) descending order sorts to corresponding disease/disease, the result that sequence is obtained represents with R and presents to user, wherein R={d|M (A, d) > 0 and r (d) < N}, r (d) represents press the sequence number of disease/disease corresponding after the descending sequence of M (A, d), and N is the constant of artificial setting.

Further, said method is further comprising the steps of:

Described disease/disease database is supplemented and upgraded.

Further, the symptom set of described effective request refers in the matching result of this request and contains disease/disease that matching degree is greater than the constant C of setting.

One group of symptom symptom corresponding with the disease of having included in system and syndrome that the present invention provides user is mated, automatically infer and cause possible disease and the syndrome of this group symptom by calculating matching degree, thereby provide relatively accurate diagnostic result to user.

One of ordinary skill in the art will appreciate that: accompanying drawing is the schematic diagram of an embodiment, the module in accompanying drawing or flow process might not be that enforcement the present invention is necessary.

One of ordinary skill in the art will appreciate that: the module in the device in embodiment can be described and be distributed in the device of embodiment according to embodiment, also can carry out respective change and be arranged in the one or more devices that are different from the present embodiment.The module of above-described embodiment can be merged into a module, also can further split into multiple submodules.

Finally it should be noted that: above embodiment only, in order to technical scheme of the present invention to be described, is not intended to limit; Although the present invention is had been described in detail with reference to previous embodiment, those of ordinary skill in the art is to be understood that: its technical scheme that still can record previous embodiment is modified, or part technical characterictic is wherein equal to replacement; And these amendments or replacement do not make the essence of appropriate technical solution depart from the spirit and scope of embodiment of the present invention technical scheme.

Claims (10)

1. the auto-check system based on symptom coupling and machine learning, is characterized in that, comprising:

Disease/disease database, for preserving known every kind of disease/disease and corresponding symptom thereof;

User interactive module, for receiving the symptom keyword set of user's input;

Symptom matching module, for mating with the symptom of described disease/disease database according to the described symptom keyword set of user's input, calculates the matching degree of described symptom keyword set and every kind of disease/disease;

Diagnostic module, for determining corresponding disease/disease according to described symptom keyword set with the matching degree of every kind of disease/disease.

2. auto-check system according to claim 1, is characterized in that, also comprises:

Vocabulary builds module, for building symptom degree of correlation vocabulary, is specially:

Obtain symptom data, wherein said symptom data comprise the symptom obtained from textbook, dictionary with the symptom set of, near synonym table, every disease/disease obtaining from described disease/disease database and ask every effectively symptom set of request of obtaining record from user;

For obtained symptom data, suppose to have two symptom x and y, the degree of association μ (x, y) of these two symptom x and y is

$\mathrm{\&mu;}(x,y)={\mathrm{\&Sigma;}}_P[\mathrm{\&rho;}\left(P\right)\&CenterDot;\frac{{\mathrm{\&Sigma;}}_{r\&Element;P}g(r,x,y)}{{\mathrm{\&Sigma;}}_{p\&Element;P}f(p,x)\&CenterDot;{\mathrm{\&Sigma;}}_{q\&Element;P}f(q,y)}]$

Wherein (P represents the judgement weight of data source P to ρ, artificially sets ρ (with near synonym table) > ρ (disease/syndrome storehouse) >=ρ (user asks record) according to expertise; R, p, q represents the each syndrome set in data source P;

wherein | p| represents the number of the symptom containing in symptom set p,

Two symptoms that the degree of association is greater than to degree of association threshold value are saved in the symptom degree of correlation vocabulary of establishment.

3. auto-check system according to claim 2, is characterized in that, described symptom matching module comprises:

Weight calculation unit, for calculate the weights W (d, x) of symptom x at disease/disease d according to following formula:

Wherein, ρ (S) represents the weight of data source S, and e represents each description unit information of diseases related/disease in data source S;

Matching degree computing unit, for calculating every disease/disease of described disease/disease database with respect to the matching degree of described symptom keyword set, is specially:

Suppose that the described symptom keyword set that user provides is combined into A, travel through each disease/disease d and corresponding symptom set σ (d) thereof in described disease/disease database;

Matching degree M (A, d) with following formula calculating disease/disease d with respect to described symptom keyword set A:

$M(A,d)=\frac{{\mathrm{\&Sigma;}}_{x\&Element;A,y\&Element;\mathrm{\&sigma;}\left(d\right)}\mathrm{\&mu;}(x,y)\&CenterDot;W(d,y)}{\left|A\right|\&CenterDot;|\mathrm{\&sigma;}\left(d\right)|};$

Wherein, | A| and | σ (d) | represent respectively the element number in set A and set σ (d);

Press M (A, d) descending order sorts to corresponding disease/disease, the result that sequence is obtained represents with R and presents to user, wherein R={d|M (A, d) > 0 and r (d) < N}, r (d) represents press the sequence number of disease/disease corresponding after the descending sequence of M (A, d), and N is the constant of artificial setting.

4. according to the auto-check system described in any one in claim 1-3, it is characterized in that, also comprise:

Update module, for supplementing and upgrading described disease/disease database.

5. auto-check system according to claim 2, is characterized in that, the symptom set of described effective request refers in the matching result of this request and contains disease/disease that matching degree is greater than the constant C of setting.

6. the automatic diagnosis method based on symptom coupling and machine learning, is characterized in that, comprises the following steps:

Receive the symptom keyword set of user's input;

Mate with the symptom in disease/disease database according to the symptom keyword set of user's input, calculate the matching degree of described symptom keyword set and every kind of disease/disease, wherein said disease/disease database is preserved known every kind of disease/disease and corresponding symptom thereof;

Determine corresponding disease/disease according to described symptom keyword set with the matching degree of every kind of disease/disease.

7. automatic diagnosis method according to claim 6, is characterized in that, further comprising the steps of before mating step according to the symptom keyword set of user's input with the symptom in disease/disease database:

Build symptom degree of correlation vocabulary, specifically comprise:

Obtain symptom data, wherein said symptom data comprise the symptom obtained from textbook, dictionary with the symptom set of, near synonym table, every disease/disease obtaining from described disease/disease database and ask every effectively symptom set of request of obtaining record from user;

For obtained symptom data, suppose to have two symptom x and y, the degree of association μ (x, y) of these two symptom x and y is

$\mathrm{\&mu;}(x,y)={\mathrm{\&Sigma;}}_P[\mathrm{\&rho;}\left(P\right)\&CenterDot;\frac{{\mathrm{\&Sigma;}}_{r\&Element;P}g(r,x,y)}{{\mathrm{\&Sigma;}}_{p\&Element;P}f(p,x)\&CenterDot;{\mathrm{\&Sigma;}}_{q\&Element;P}f(q,y)}]$

ρ (P) represents the judgement weight of data source P, artificially sets ρ (with near synonym table) > ρ (disease/syndrome storehouse) >=ρ (user asks record) according to expertise; R, p, q represents the each syndrome set in data source P;

wherein | p| represents the number of the symptom containing in symptom set p,

Two symptoms that the degree of association is greater than to degree of association threshold value are saved in the symptom degree of correlation vocabulary of establishment.

8. automatic diagnosis method according to claim 7, it is characterized in that, the described symptom keyword set according to user's input is mated with the symptom in disease/disease database, and the matching degree step of calculating described symptom keyword set and every kind of disease/disease comprises:

Calculating the weights W (d, x) of symptom x in disease/disease d is

Wherein, ρ (S) represents the weight of data source S, and e represents each description unit information of diseases related/disease in data source S;

Suppose that the described symptom keyword set that user provides is combined into A, travel through each disease/disease d and corresponding symptom set σ (d) thereof in described disease/disease database;

Matching degree M (A, d) with following formula calculating disease/disease d with respect to described symptom keyword set A

$M(A,d)=\frac{{\mathrm{\&Sigma;}}_{x\&Element;A,y\&Element;\mathrm{\&sigma;}\left(d\right)}\mathrm{\&mu;}(x,y)\&CenterDot;W(d,y)}{\left|A\right|\&CenterDot;|\mathrm{\&sigma;}\left(d\right)|};$

Wherein, | A| and | σ (d) | represent respectively the element number in set A and set σ (d);

Press M (A, d) descending order sorts to corresponding disease/disease, the result that sequence is obtained represents with R and presents to user, wherein R={d|M (A, d) > 0 and r (d) < N}, r (d) represents press the sequence number of disease/disease corresponding after the descending sequence of M (A, d), and N is the constant of artificial setting.

9. according to the automatic diagnosis method described in any one in claim 6-8, it is characterized in that, further comprising the steps of:

Described disease/disease database is supplemented and upgraded.

10. automatic diagnosis method according to claim 7, is characterized in that, the symptom set of described effective request refers in the matching result of this request and contains disease/disease that matching degree is greater than the constant C of setting.